Forest smoke and fire detection method based on target detection, storage medium and equipment

A target detection and fire detection technology, which is applied to forest fire alarms, fire alarms that rely on radiation effects, fire alarms, etc., can solve the problem of inability to accurately detect the specific location of fireworks and forest fires. Problems such as the location of the occurrence and the difficulty for firefighters to quickly reach the designated location, etc., to achieve the effects of improved pyrotechnic recognition rate, large field of view, and good robustness

Pending Publication Date: 2020-04-07
NANJING ENBO TECH
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

And when the method of image classification is used, the specific location where the fireworks occur cannot be accurately detected. Therefore, when a fire breaks out in the forest, it is difficult for firefighters to quickly reach the designated location to fight the forest fire.
[0005] In summary, the existing forest fire detection methods are easily affected by the external environment, the accuracy of detection is low, and they cannot accurately detect the location of forest fire, which affects forest fire fighting.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Forest smoke and fire detection method based on target detection, storage medium and equipment
  • Forest smoke and fire detection method based on target detection, storage medium and equipment
  • Forest smoke and fire detection method based on target detection, storage medium and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention will be further described below in conjunction with embodiment and accompanying drawing.

[0038] combine figure 1 As shown, the specific implementation process of the forest firework detection method based on target detection of the present invention is described in detail. The method passes through the forest firework video monitoring system or removes the forest firework image, and then uses the neural network model to perform forward calculation on the forest firework image. Obtain the forest firework area and its confidence level, and then judge the forest fire situation according to the confidence level. The main part of this method lies in the construction and training of the neural network model. according to figure 1 In the dotted box part, the process of building and training the neural network model is as follows:

[0039] S1: Establish a sample dataset of forest fireworks images

[0040] Based on the forest pyrotechnic monitoring syste...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a forest smoke and fire detection method based on target detection, a storage medium and equipment, and belongs to the field of computer vision target detection and forest smoke and fire video monitoring. The method includes: acquiring a forest smoke and fire image; performing forward calculation on the forest smoke and fire image by adopting a neural network model comprising a multi-level network to obtain a smoke and fire area and a confidence coefficient thereof; comparing the confidence coefficient with a confidence coefficient threshold value, and when the confidence coefficient is larger than the confidence coefficient threshold value, considering that a fire occurs, and when the confidence coefficient is smaller than or equal to the confidence coefficient threshold value, considering that no fire occurs. According to the invention, the network model with high precision is obtained by using small samples, the influence of the environment is small, the detection accuracy is high, false alarm is not easy to occur, and the bounding box of the smoke and fire area can be obtained, so that the fire occurrence position can be judged, and firefighters can quickly arrive at the fire area to extinguish the fire. Meanwhile, the disclosed storage medium and equipment can be directly deployed and used for forest smoke and fire detection.

Description

technical field [0001] The invention belongs to the fields of computer vision target detection and forest firework video monitoring, and in particular relates to a forest firework detection method, storage medium and equipment based on target detection. Background technique [0002] Forest fires are a common natural disaster. The casualties and property losses caused by forest fires in the world are huge every year. For example, the recent California forest fires, Indonesian forest fires, and Amazon forest fires have not only caused a large number of Casualties and property losses have also caused serious impacts on the natural environment. In the fire fighting process, a lot of manpower and material resources have been invested, but it is still difficult to effectively extinguish forest fires. Forest fires have caused serious threats to the survival of humans and animals. my country is a country with many forests. Every year, tens of thousands of casualties and hundreds of...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G08B17/00G08B17/12
CPCG08B17/005G08B17/125G06N3/044G06N3/045G06F18/24G06F18/214
Inventor 张广铭洪刚俊陆勇
Owner NANJING ENBO TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products